Osteosarcoma (OS) is the most common primary malignant bone tumour in humans and dogs. Although medicine has made dramatic progress in treating osteosarcoma by surgery, with chemotherapy given before and after surgery, drug resistance and highly metastatic spread are often responsible for the failure of current therapies. Thus, more effective therapeutic approaches for treating osteosarcoma are needed. Previous results from our laboratory and others had shown that the basic-leucine zipper (bLZip) containing transcription factor, Zhangfei/CREBZF is a potent inhibitor of a variety of other transcription factors and has a dramatic effect on the growth of several cancer cell lines, including dog OS and human medulloblastoma cells. The objective of the studies described in this thesis was to determine the molecular mechanisms by which Zhangfei exerts its effect on dog and human OS cells. Several stressors in the microenvironment of cancer cells directly or indirectly perturb the endoplasmic reticulum (ER), which then activates the Unfolded Protein Response (UPR). The UPR modulates the effects of stress and allows tumours to survive, develop, metastasize and escape therapy. The UPR is regulated by three bLZip transcription factors—ATF6, ATF4 and Xbp1s. Since Zhangfei inhibits Luman/CREB3, a bLZip structurally similar to and closely related to ATF6 and ATF4, I initially focused my efforts on this pathway. I hypothesized that Zhangfei interacts with UPR-related bLZip transcription factors and inhibits their ability to activate the UPR signaling pathways, thereby suppressing the growth of cancer cells and increasing their susceptibility to ER stressors. To test this hypothesis, we monitored cell growth as well as levels of UPR gene transcripts and proteins in several dog and human osteosarcoma cell lines infected with adenovirus vectors expressing Zhangfei, and studied the interactions between Zhangfei and the UPR-mediator, Xbp1s. The results showed that the ectopic expression of Zhangfei in cell lines derived from dog osteosarcomas potently suppressed cell growth and inhibited their ability to activate the UPR. Further studies demonstrated that Zhangfei inhibited the UPR, at least partially, by binding to Xbp1s and suppressing its ability to activate transcription from a promoter containing unfolded protein response elements (UPRE). The leucine zipper of Zhangfei was required for this interaction, which led to the subsequent proteasomal degradation of Xbp1s. However, we also found that the effects of Zhangfei were not universal. While Zhangfei had a profound effect on the growth and UPR in some OS cell lines, it either had only a partial effect, or no effect on others. This suggested that susceptibility (or resistance) to Zhangfei may be an inherent property of OS cell lines. Since the suppressive effects of Zhangfei were not universal, and it had no obvious effects on untransformed cells and some cancer cell lines, I proposed that Zhangfei mediates its effect on cell growth and the UPR through an intermediary that is either not induced or is defective in cells that are unaffected by Zhangfei. I found that this intermediary was the tumour suppressor protein p53. The inhibitory effects of Zhangfei were only observed in the wild-type p53 expressing OS cell line U2OS while Zhangfei had no effect on the p53-null OS cell line MG63. In cells with functional p53, the ectopic expression of Zhangfei caused it to displace the ubiquitin ligase mdm2 and stabilize p53. Suppression of p53 by siRNA partially inhibited the effects of Zhangfei on the UPR and cell growth. In contrast, OS cells lacking functional p53 could be made to respond to Zhangfei if they were transfected to express wild-type p53. These results explain why Zhangfei has a profound effect on some cancer cells while having no obvious effect on others. I also characterized the interaction of Zhangfei and p53 by mapping the interacting domains on both proteins, showing that the bLZip domain of Zhangfei and the N-terminal transactivation domain (NTD) of p53 were required for their interactions. My findings reveal the profoundly inhibitory effects of Zhangfei on OS growth and the UPR, a stress-response known to promote tumour survival. I also show how Zhangfei may exert its effects. My work suggests an alternative modality for the therapy of certain types of OS, and perhaps other tumours with functional p53.

Medulloblastoma cells do not contain detectable amounts of the bZIP protein Zhangfei. However, previous work has shown that expression of this protein in cells of the ONS-76 line, derived from a human medulloblastoma, causes the cells to stop growing and develop processes that resemble neuritis (a characteristic of differentiated neurons). Zhangfei-expressing cells eventually die. My objective was to determine the molecular mechanisms by which Zhangfei influences ONS-76 cells. My strategy was to infect ONS-76 cells with adenovirus vectors expressing either Zhangfei or the control E. coli protein â-galactosidase (LacZ) and then to compare the following parameters in Zhangfei and LacZ-expressing cells: a) markers of apoptosis, autophagy and macropinocytosis (the three main pathways of cell death); b) transcripts for genes involved in neurogenesis and apoptosis; c) phosphorylation of peptide targets of selected cellular protein kinases; and d) active transcription factors. Zhangfei-expressing cells appeared to succumb to apoptosis as determined by the expression of phosphatidylserine on the cell surface and intensity of nuclear staining with the DNA dye Hoechst. Increased staining for autophagic vesicles and upregulated expression of autophagy response genes in these cells indicated that they were undergoing autophagy, possibly associated with apoptosis. My analysis of steady-state transcripts for genes involved in apoptosis and neurogenesis and functional protein kinases in Zhangfei-expressing cells indicated that the mitogen-activated protein kinase (MAPK) pathway was active in these cells. In addition, I found that the transcription factor Brn3a as well as factors implicated in differentiation were also active. These observations led me to hypothesize that Zhangfei enhances the expression of Brn3a, a known inducer of TrkA, the high-affinity receptor for nerve growth factor (NGF). TrkA then binds in an autocrine manner to NGF, triggering the MAPK pathway and leading to differentiation of ONS-76 cells into neuron and glia-like cells, eventually bringing about cell death by apoptosis and autophagy. I tested this hypothesis by showing that Zhangfei could enhance transcription from the isolated Brn3a promoter, that ONS-76 cells produce NGF as detected in a bioassay, and that antibodies against NGF and inhibitors of TrkA and selected components of the MAPK pathway could partially restore the growth of Zhangfei-expressing ONS-76 cells. My work supports previous work highlighting the importance of NGF-TrkA signaling in the outcome of medulloblastomas and shows how Zhangfei is able to trigger this pathway.

Medulloblastoma cells do not contain detectable amounts of the bZIP protein Zhangfei. However, previous work has shown that expression of this protein in cells of the ONS-76 line, derived from a human medulloblastoma, causes the cells to stop growing and develop processes that resemble neuritis (a characteristic of differentiated neurons). Zhangfei-expressing cells eventually die. My objective was to determine the molecular mechanisms by which Zhangfei influences ONS-76 cells. My strategy was to infect ONS-76 cells with adenovirus vectors expressing either Zhangfei or the control E. coli protein â-galactosidase (LacZ) and then to compare the following parameters in Zhangfei and LacZ-expressing cells: a) markers of apoptosis, autophagy and macropinocytosis (the three main pathways of cell death); b) transcripts for genes involved in neurogenesis and apoptosis; c) phosphorylation of peptide targets of selected cellular protein kinases; and d) active transcription factors. Zhangfei-expressing cells appeared to succumb to apoptosis as determined by the expression of phosphatidylserine on the cell surface and intensity of nuclear staining with the DNA dye Hoechst. Increased staining for autophagic vesicles and upregulated expression of autophagy response genes in these cells indicated that they were undergoing autophagy, possibly associated with apoptosis. My analysis of steady-state transcripts for genes involved in apoptosis and neurogenesis and functional protein kinases in Zhangfei-expressing cells indicated that the mitogen-activated protein kinase (MAPK) pathway was active in these cells. In addition, I found that the transcription factor Brn3a as well as factors implicated in differentiation were also active. These observations led me to hypothesize that Zhangfei enhances the expression of Brn3a, a known inducer of TrkA, the high-affinity receptor for nerve growth factor (NGF). TrkA then binds in an autocrine manner to NGF, triggering the MAPK pathway and leading to differentiation of ONS-76 cells into neuron and glia-like cells, eventually bringing about cell death by apoptosis and autophagy. I tested this hypothesis by showing that Zhangfei could enhance transcription from the isolated Brn3a promoter, that ONS-76 cells produce NGF as detected in a bioassay, and that antibodies against NGF and inhibitors of TrkA and selected components of the MAPK pathway could partially restore the growth of Zhangfei-expressing ONS-76 cells. My work supports previous work highlighting the importance of NGF-TrkA signaling in the outcome of medulloblastomas and shows how Zhangfei is able to trigger this pathway.

The unfolded protein response (UPR) is a well conserved mechanism in eukaryotes that protects organisms from the damaging effects of endoplasmic reticulum (ER) stresses. Activation of the UPR will lead to two outcomes. It first attempts to restore cellular functions by enhancing protein folding capacity, inhibiting protein synthesis and promoting degradation of harmful proteins (the pro-survival pathway). However, if the stressful conditions are prolonged or severe, apoptosis will be induced (the pro-apoptotic pathway). The present study suggests that an isoform of the cellular protein Zhangfei (ZF, CREBZF) is linked to the pro-apoptotic pathway in the UPR by using DNA, protein and cell viability analyses. This isoform is known as the short-tail ZF (stZF). We demonstrated that stZF can be induced by prolonged ER stress. The protein of stZF is stable under ER stress and it has the ability to promote programmed cell death in the early stage of apoptosis through the induction of CHOP, a protein that plays a key role in the pro-apoptotic pathway of the UPR.

Herpes simplex viruses (HSV) establish latent infections in sensory neurons of their host and are maintained in this state by little understood mechanisms that, at least in part, are regulated by signalling through nerve growth factor (NGF) and its receptor tropomyosin related kinase, trkA. Previous studies have demonstrated that Zhangfei is a transcriptional factor that is expressed in differentiated neurons and is thought to influence HSV replication and latency. Zhangfei, like the HSV trans-activator VP16 and Luman, binds the ubiquitous nuclear protein host cell factor (HCF) inhibiting the ability of VP16 and Luman to initiate HSV replication. <p>Recently, Brn3a, another neuronal factor thought to influence HSV latency and reactivation was found to possess an HCF-binding domain and could potentially require HCF for activity. The neuronal POU IV domain protein, Brn3a, among its many regulatory functions has been described as an enhancer of the NGF receptor trkA, during development in mouse. I therefore investigated the possible link between Brn3a, TrkA, NGF signaling, HCF, Zhangfei and HSV-1 latency and reactivation. I hypothesized that Zhangfei would also suppress the ability of Brn3a to activate the expression of TrkA and that this would have an impact on NGF-TrkA signaling and, consequently on HSV-1 reactivation from latency.<p>My first study determined which Brn3a/trkA promoter interactions were important for trkA transcription. I constructed a plasmid that contains 1043 base pairs of genomic sequences that extend from 30 nucleotides upstream of trkA coding region. In contrast to previous data, a short 190 bp region that lies proximal to the trkA initiation codon was sufficient for Brn3a trans-activation in NGF-differentiated PC12, Vero and human medulloblastoma cells. At least two portions of the 190 bp fragment bind to Brn3a. In addition, Brn3a increased endogenous levels of trkA transcripts in PC12 cells and initiated trkA expression in medulloblastoma cells, which normally do not express trkA. <p>The second step was to determine the effects of HCF and Zhangfei association with Brn3a on trkA trans-activation. I found that Brn3a required HCF for activating the trkA promoter and that Zhangfei has a suppressive effect over Brn3a-trkA activation in non-neuronal cells. In sympathetic neuron-like NGF-treated PC12 cells, Zhangfei did not suppress the ability of Brn3a to activate the TrkA promoter, however, Zhangfei was able capable of inducing the expression of TrkA in the absence of Brn3a. Both Brn3a and Zhangfei induced the expression of endogenous trkA in PC12 cells.<p>Since Vero and PC12 cells are not from human origin I wanted to examine the ability of Zhangfei to induce trkA transcription in medulloblastoma cells, that because of its tumor nature do not express trkA. TrkA transfections in these cells have shown to drive them to cell arrest or apoptosis. Since Zhangfei is not express in medulloblastoma tumors I then used ONS-76 medulloblastoma cells as a model to determine Zhangfeis envolvement in the NGF-trkA signaling pathway.<p> I show herein that in ONS-76 medulloblastoma cells resveratrol, an inducer of apoptosis and differentiation, increased the expression of Zhangfei and trkA as well as Early Growth Response Gene 1 (Egr1), a gene normally activated by NGF-trkA signalling. ONS-76 cells stop growing soon after treatment with resveratrol and a portion of the cell undergo apoptosis. While the induction of Zhangfei in resveratrol-treated cells was modest albeit consistent, the infection of actively growing medulloblastoma cells with an adenovirus vector expressing Zhangfei mimicked the effects of resveratrol. Zhangfei activated the expression of trkA and Egr1 and caused these cells to display markers of apoptosis. The phosphorylation of Erk1, an intermediate kinase in the NGF-trkA signaling critical for differentiation, was observed in Zhangfei infected cells, supporting the hypothesis that Zhangfei is a mediator of trkA-NGF signaling in theses cells leading either to differentiation or apoptosis. Binding of HCF by Zhangfei did not appear to be required for this effect as a mutant of Zhangfei incapable of binding HCF was also able to induce the expression of trkA and Egr1. <p>In in vivo and in vitro models of HSV-1 latency, the virus reactivates when NGF supply to the neuron is interrupted. Based on the above evidence Zhangfei, in HSV-1 latently infected neurons, would have the ability to prolong a state of latency by inducing trkA expression allowing the activation of NGF-trkA signaling pathway. Since NGF is produced by many cell types it is possible that reactivation is triggered not by a decrease in NGF but by a down-regulation of TrkA expression.Therefore, if Zhangfei expression is suppress the trkA signaling could be interrupted or shifted towards apoptosis signaling, this would allow neuronal HCF-binding proteins like Luman, which can activate HSV IE expression, to initiate HSV IE expression and subsequently viral replication.

Herpes simplex viruses (HSV) establish latent infections in sensory neurons of their host and are maintained in this state by little understood mechanisms that, at least in part, are regulated by signalling through nerve growth factor (NGF) and its receptor tropomyosin related kinase, trkA. Previous studies have demonstrated that Zhangfei is a transcriptional factor that is expressed in differentiated neurons and is thought to influence HSV replication and latency. Zhangfei, like the HSV trans-activator VP16 and Luman, binds the ubiquitous nuclear protein host cell factor (HCF) inhibiting the ability of VP16 and Luman to initiate HSV replication. <p>Recently, Brn3a, another neuronal factor thought to influence HSV latency and reactivation was found to possess an HCF-binding domain and could potentially require HCF for activity. The neuronal POU IV domain protein, Brn3a, among its many regulatory functions has been described as an enhancer of the NGF receptor trkA, during development in mouse. I therefore investigated the possible link between Brn3a, TrkA, NGF signaling, HCF, Zhangfei and HSV-1 latency and reactivation. I hypothesized that Zhangfei would also suppress the ability of Brn3a to activate the expression of TrkA and that this would have an impact on NGF-TrkA signaling and, consequently on HSV-1 reactivation from latency.<p>My first study determined which Brn3a/trkA promoter interactions were important for trkA transcription. I constructed a plasmid that contains 1043 base pairs of genomic sequences that extend from 30 nucleotides upstream of trkA coding region. In contrast to previous data, a short 190 bp region that lies proximal to the trkA initiation codon was sufficient for Brn3a trans-activation in NGF-differentiated PC12, Vero and human medulloblastoma cells. At least two portions of the 190 bp fragment bind to Brn3a. In addition, Brn3a increased endogenous levels of trkA transcripts in PC12 cells and initiated trkA expression in medulloblastoma cells, which normally do not express trkA. <p>The second step was to determine the effects of HCF and Zhangfei association with Brn3a on trkA trans-activation. I found that Brn3a required HCF for activating the trkA promoter and that Zhangfei has a suppressive effect over Brn3a-trkA activation in non-neuronal cells. In sympathetic neuron-like NGF-treated PC12 cells, Zhangfei did not suppress the ability of Brn3a to activate the TrkA promoter, however, Zhangfei was able capable of inducing the expression of TrkA in the absence of Brn3a. Both Brn3a and Zhangfei induced the expression of endogenous trkA in PC12 cells.<p>Since Vero and PC12 cells are not from human origin I wanted to examine the ability of Zhangfei to induce trkA transcription in medulloblastoma cells, that because of its tumor nature do not express trkA. TrkA transfections in these cells have shown to drive them to cell arrest or apoptosis. Since Zhangfei is not express in medulloblastoma tumors I then used ONS-76 medulloblastoma cells as a model to determine Zhangfeis envolvement in the NGF-trkA signaling pathway.<p> I show herein that in ONS-76 medulloblastoma cells resveratrol, an inducer of apoptosis and differentiation, increased the expression of Zhangfei and trkA as well as Early Growth Response Gene 1 (Egr1), a gene normally activated by NGF-trkA signalling. ONS-76 cells stop growing soon after treatment with resveratrol and a portion of the cell undergo apoptosis. While the induction of Zhangfei in resveratrol-treated cells was modest albeit consistent, the infection of actively growing medulloblastoma cells with an adenovirus vector expressing Zhangfei mimicked the effects of resveratrol. Zhangfei activated the expression of trkA and Egr1 and caused these cells to display markers of apoptosis. The phosphorylation of Erk1, an intermediate kinase in the NGF-trkA signaling critical for differentiation, was observed in Zhangfei infected cells, supporting the hypothesis that Zhangfei is a mediator of trkA-NGF signaling in theses cells leading either to differentiation or apoptosis. Binding of HCF by Zhangfei did not appear to be required for this effect as a mutant of Zhangfei incapable of binding HCF was also able to induce the expression of trkA and Egr1. <p>In in vivo and in vitro models of HSV-1 latency, the virus reactivates when NGF supply to the neuron is interrupted. Based on the above evidence Zhangfei, in HSV-1 latently infected neurons, would have the ability to prolong a state of latency by inducing trkA expression allowing the activation of NGF-trkA signaling pathway. Since NGF is produced by many cell types it is possible that reactivation is triggered not by a decrease in NGF but by a down-regulation of TrkA expression.Therefore, if Zhangfei expression is suppress the trkA signaling could be interrupted or shifted towards apoptosis signaling, this would allow neuronal HCF-binding proteins like Luman, which can activate HSV IE expression, to initiate HSV IE expression and subsequently viral replication.